Hi,
according to autoconfigure I have increased parameters on DB2 ESE v8.2 fp11 on Linux.

I have changed the following parameters:
db2 UPDATE DBM CFG USING FCM_NUM_RQB 512
db2 UPDATE DBM CFG USING MAX_QUERYDEGREE 1
db2 UPDATE DBM CFG USING NUM_POOLAGENTS 400
db2 UPDATE DBM CFG USING SHEAPTHRES 13500
db2 UPDATE DB CFG FOR mydb USING APPGROUP_MEM_SZ 11560
db2 UPDATE DB CFG FOR mydb USING CATALOGCACHE_SZ 812
db2 UPDATE DB CFG FOR mydb USING DBHEAP 3592
db2 UPDATE DB CFG FOR mydb USING LOCKLIST 4400
db2 UPDATE DB CFG FOR mydb USING LOGBUFSZ 124
db2 UPDATE DB CFG FOR mydb USING LOGFILSIZ 1024
db2 UPDATE DB CFG FOR mydb USING LOGSECOND 155
db2 UPDATE DB CFG FOR mydb USING MAXAPPLS 112
db2 UPDATE DB CFG FOR mydb USING MAXLOCKS 56
db2 UPDATE DB CFG FOR mydb USING NUM_IOCLEANERS 2
db2 UPDATE DB CFG FOR mydb USING NUM_IOSERVERS 6
db2 UPDATE DB CFG FOR mydb USING PCKCACHESZ 975
db2 UPDATE DB CFG FOR mydb USING SOFTMAX 120
db2 UPDATE DB CFG FOR mydb USING SORTHEAP 192
db2 UPDATE DB CFG FOR mydb USING UTIL_HEAP_SZ 36419
db2 ALTER BUFFERPOOL IBMDEFAULTBP SIZE 109260

And system admin has told me that number of Linux processes has doubled compare to number of processes before changing dbm/db/bufferpool parameters - see attached picture.

Which of dbm/db/bufferpool settings has increased number of processes?

Thanks,
Grofaty

I cannot see where a heap size change would affect the number of processes.
So you can eliminate everything but FCM_NUM_RQB, MAX_QUERYDEGREE, NUM_POOLAGENTS, MAXAPPLS, NUM_IOCLEANERS, NUM_IOSERVERS, and MAXLOCKS.

MAXLOCKS is a percentage, so you can eliminate that. I think setting max_querydegree to one would reduce agents since there will no longer an parallelism for queries.

What was the values of the remaining parameters before the change?

Andy

I left my crystal ball at home today, unfortunately, so I'll wait until I see the list of processes before I give any advice.

Maybe more agents because the pool is probably bigger? Maybe more iocleaners/ioservers because that number is bigger? Who knows...